AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Are LLMs Really Not Knowledgeable? Mining the Submerged Knowledge in LLMs' Memory

Xingjian Tao, Yiwei Wang, Yujun Cai, Zhicheng Yang, Jing Tang

ICLR 2026 · Large language models › Retrieval and knowledge · approach: Knowledge editing in LLMs · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
measure Hits@5 instead of Hits@1 (top-k logit retention) (ablation)LLaMA3-8b · DBpediaHits@k: 17.2 -> 57.9Improved
increase k to 50 (ablation)LLaMA3-8b · DBpedia headHits@k: 48.3 (k=5) -> 83.4 (k=50) -> 90.5 (k=100)Improved
scale LLaMA3 from 8b to 70b (ablation)LLaMA3 · DBpedia head, k=100Hits@100: 90.5 -> 92.1No clear effect
scale LLaMA3 from 8b to 70b (ablation)LLaMA3 · GoodReads head, k=100Hits@100: 67.8 -> 44.2Hurt
newer model generation (LLaMA3 vs LLaMA2)LLaMA · DBpedia head, k=100Hits@100: 70.5 (LLaMA2-70b) -> 90.5 (LLaMA3-8b)Improved
filter unsure-related tokens during decoding (ablation)LLaMA3-70b · DBpedia headaccuracy (%): 11.2 -> 23.0Improved
filter unsure-related tokens during decoding (ablation)Mistral-7b · DBpedia headaccuracy (%): 16.5 -> 16.7No clear effect
filter unsure-related tokens during decoding (ablation)LLaMA2-70b · IMDB headaccuracy (%): 25.2 -> 25.2No clear effect
filter unsure-related tokens during decoding (ablation)LLaMA3-8b · IMDB tailaccuracy (%): 2.2 -> 2.7No clear effect
open-domain vs domain-specific dataLLaMA3-8b · Hits@100 DBpedia vs IMDB headHits@100: 90.5 (DBpedia) vs 69.7 (IMDB)Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy